Journal of Industrial and Engineering Chemistry 20 (2014) 1517–1522

Preparation and characterization of polyvinyl chloride basednanocomposite nanofiltration-membrane modified by iron oxidenanoparticles for lead removal from water

A. Gholami a, A.R. Moghadassi a,*, S.M. Hosseini a, S. Shabani a, F. Gholami b

a Department of Chemical Engineering, Faculty of Engineering, Arak University, Arak 38156-8-8349, Iranb Department of Chemistry, Faculty of Science, Arak University, Arak 38156-8-8349, Iran

A R T I C L E I N F O

Article history:

Received 6 February 2013

Accepted 23 July 2013

Available online 6 August 2013

Keywords:

Nanocomposite membrane

Nanofiltration

Membranes hydrophilicity

Iron oxide nanoparticle

Lead removal

A B S T R A C T

In this research (polyvinyl chloride-blend-cellulose acetate/iron oxide nanoparticles) nanocomposite

membranes were prepared by casting technique to lead removal from wastewaters. The effect of blend

ratio of polymer binder (PVC to CA) and Fe3O4 nanoparticles concentration on physico-chemical

characteristics of membranes were studied. Water permeability and ionic rejection tests, water content

and mechanical properties measurements and SEM analysis were carried out in membranes

characterizations. Obviously, modified membrane containing 10 wt% CA and 0.1 wt% Fe3O4 nanopar-

ticles showed better performance in lead removal compared to other modified membranes and also

pristine ones.

� 2013 The Korean Society of Industrial and Engineering Chemistry. Published by Elsevier B.V. All rights

reserved.

Contents lists available at ScienceDirect

Journal of Industrial and Engineering Chemistry

jou r n al h o mep ag e: w ww .e lsev ier . co m / loc ate / j iec

1. Introduction

Wastewaters which contain heavy metals have attracted globalattention because of their harmful effects on the environment andhuman health, certainly to be a risk for human beings [1]. Lead isone of the heavy metal that is non-essential and occurs naturally inthe environment. But some physical and chemical properties oflead such as ductility, softness, malleability, poor conductivity andresistance to corrosion have led to wide use of this metal sinceancient times in variety of applications. Today, the maximumconcentrations found in environment are the result of humanactions, while it is demonstrated that lead can cause dangeroushealth effects, even at trace level, such as nephrotoxicity,neurotoxicity and adverse effects on the hematological andcardiovascular systems [1,2]. So, it is very important to removeLead from wastewaters effectively before their discharge into theenvironment. Several treatment techniques such as adsorption,electrochemical oxidation, coagulation-flocculation, and mem-brane filtration are used to eliminate pollutants from wastewater.Membrane process has verified to be very attractive and effectivefor the treatment of effluents. The main advantage of membrane-based technologies is that they do not need the addition of

* Corresponding author. Tel.: +98 861 262 5422; fax: +98 861 262 5423.

E-mail addresses: A_Moghadassi@yahoo.com, A-Moghadassi@araku.ac.ir

(A.R. Moghadassi).

1226-086X/$ – see front matter � 2013 The Korean Society of Industrial and Engineer

http://dx.doi.org/10.1016/j.jiec.2013.07.041

chemicals [3]. One of membrane techniques is Nano filtration (NF),that is becoming progressively universal in concentration andpurification, pharmaceutical and chemical industries, watertreatment and desalination [4,5]. NF has some good advantages,such as relatively low investment, low energy consumption, highpermeation flux and singular separation capability for ions ofdifferent valences [6].

In NF membranes, two mechanisms affect the rejectionbehavior of solutes, surface charge and sieving, that are usuallycharacterized using high rejection of bivalent ions, lower rejectionof monovalent ions, and higher fluxes than reverse osmosis (RO)membranes [5,7,8]. NF membranes are increasingly known as oneof the best procedures for water treatment and production ofportable water [9].

Among the various polymeric materials which are used inpreparation of membranes, polyvinyl chloride (PVC) is flexible anddurable polymer with suitable biological and chemical resistance[10–12]. CA is also one of the most applicable polymers infabrication of membranes due to its high hydrophilicity, goodtoughness, high biocompatibility, good desalting, high potentialflux, good resistance to chlorine and solvent and it is cheap [13].Use of these polymers (PVC/CA) and their blends as membranebinders can dedicate special selective characteristics into theprepared membranes and improve the separation properties ofthem [14].

One of the best modification techniques for enhancingmembrane performance is blending with inorganic nanoparticles

ing Chemistry. Published by Elsevier B.V. All rights reserved.

Table 1Structure of used polymers.

Polyvinyl chloride (PVC)

Cellulose acetate (CA)

Table 2Composition of casting solution for the fabrication of homemade membranes.

Sample no. Used polymer Solvent (wt%) THF

PVC (%w/w) CA (%w/w) Fe3O4 (%w/w)

1 100 0 0 88

2 97.5 2.5 0 88

3 95 5 0 88

4 92.5 7.5 0 88

5 90 10 0 88

6 90 10 0.01 88

7 90 10 0.1 88

8 90 10 1 88

A. Gholami et al. / Journal of Industrial and Engineering Chemistry 20 (2014) 1517–15221518

such as TiO2 [5], Al2O3 [15,16], bentonite [17], and carbonnanotubes [3,5,6,10–31]. Inserting inorganic nanoparticles in themembrane matrix can improve the strength and stiffness, thehydrophilicity, the water permeability and rejection and theantifouling properties of polymer based Nano composite mem-branes [5,18–21]. Among these, nano-sized ferric oxides are usedfor the removal of heavy metals from aqueous systems [1]. This isbecause of some good features such as excellent thermal andchemical stability, magnetic performance and good biodegrada-tion and biocompatibility. These features caused wide use of thisparticle in different separation techniques such as preparingmagnetic targeting medicine and magnetic fluids, preparingmagnetic microspheres for the quick separation of cellular,biomedical and bioengineering products [22].

The aim of this research is fabrication of novel nanofiltrationmembrane combined with absorbents particles by using nano-sized ferric oxides in membrane matrix to increase in lead removalefficiency from water. For the purpose, polyvinyl chloride-blend-cellulose membrane was prepared by solution casting technique.Nano-sized ferric oxide was also used as adsorbent in variousconcentrations to achieve nanocomposite nanofiltration mem-brane with high rejection and flux. The effect of blend ratio ofpolymer binder (PVC to CA) and Fe3O4 nanoparticles concentrationon physico-chemical characteristics of membranes were studied.

2. Experimental

2.1. Materials

Polyvinyl chloride (PVC, Bandar Imam Petroleum Company,grade S-7054) and cellulose acetate (CA, Acros company,MW = 100,000 g/mol) as polymers – their structures presentedin Table 1 – and tetrahydrofuran (THF, DAE Jung) as solvent andFe3O4 nanoparticles (MW = 213.53 g/mol, APS = 60 nm, SSA >

55 m2/g, purification = 99.2%) as a adsorbent and lead(yy)nitrate(Pb(NO3)2, Fulka company, MW = 331.2 g/mol, analysis number:334371.1 41497) and distilled water from Alborz company wereused throughout this study.

2.2. Membrane preparation

The asymmetric membranes were prepared by phase inversionmethod. The solution containing Polyvinyl Chloride (PVC) andCellulose Acetate (CA) as polymer, Tetrahydrofuran (THF) assolvent and iron oxide nanoparticles was shaken for 24 h at 60 8C[22] (according to Table 2) and cast on a clean glass plate atambient temperature by manual casting knife with 150 mmthickness. The membrane surface was exposed to air at ambienttemperature (about 26 8C) for free-convective solvent evaporation.

After a 60 s delay membranes were immersed into distillatedwater bath with the temperature about 26 8C. The polymeric skinlayer was taken out after 15 min and then immersed in distillatedwater for 24 h to ensure complete elimination of residual solvent.The final membranes thickness was also measured by a digitalcaliper device (Electronic outside Micrometer, IP54 model OLR)between 135 and 150 mm.

To achieve final membrane two steps must be carried out: First,optimization on CA concentration to obtain membrane withmaximum rejection and flux, and second step, investigation effectof various concentration of the nanoparticle adsorbent to improverejection of lead. Two stages are presented by different colors inTable 2.

2.3. Membrane characterization

2.3.1. Permeability and rejection

For characterization permeability of the home-made nanofil-tration membranes, a dead-end system was designed. This systemconsists a filtration cell with a stirrer in it and a nitrogen gascylinder. Model of Filtration cell is 8200 Millipore Co. with innerdiameter of 40 mm and volume capacity of 150 mL and effectivearea of the membrane was 12.566 cm2. Fluid that reserved in cellwas pressed by nitrogen gas. Fig. 1 shows symbolic view ofapparatus.

The water flux (Jw) that resulted by this pressure was calculatedby the following equation:

Jw ¼V

ADt(1)

V (L) is volume of permeated water from membrane that iscollected in a graduated cylinder and A (m2) is effective area of themembrane and t (h) is the time in which permeate is collected inthe graduated cylinder.

Fig. 2. The effect of CA concentration (2.5, 5, 7.5, and 10) on flux.

Fig. 3. The effect of CA concentration (2.5, 5, 7.5, and 10) on rejection.

Fig. 1. The dead end filtration setup for evaluation membrane performance: 1,

nitrogen cylinder; 2, pressure gage; 3, nitrogen entrance; 4, feed entrance; 5,

reservoir; 6, permeate; 7, collector.

A. Gholami et al. / Journal of Industrial and Engineering Chemistry 20 (2014) 1517–1522 1519

For measurement of Pb rejection following equation wasused:

R ¼ 1 � Cp

Cf

� �� 100% (2)

where Cp and Cf are concentrations of lead ions in permeate andfeed, respectively. Concentrations of Pb ions in permeate and feedwere measured by electrical conductivity (Digital conduct-meter,Jenway, Model: 4510).

2.3.2. Scanning electron microscopy (SEM) analysis

The SEM analysis provides the visual data of the top surfaceand also cross-sectional morphology of the membrane. This testwas done by Kyky (EM-3200). The unused membrane sampleswere divided into smaller size samples. The samples weresubmerged in liquid nitrogen for 60–70 s to freeze membranes.The frozen pieces were broken and kept in ambient temperaturefor drying. The dried pieces were silver-coated for producingelectric conductivity. Then, they were viewed with the micro-scope and all details presented in pictures.

2.3.3. Water content

The water content test reveals membrane hydrophilic propertyand measures the difference between wet and dry weight of themembrane. Eq. (3) shows how to calculate water content values[24]:

WC% ¼ðWwet � WdryÞ

Wwet� 100 (3)

Where Wwet is weight of wet membrane (after 48 h immersion indistillated water) and Wdry is weight of dry membrane which driedin oven until the constant weight was reached.

2.3.4. Mechanical strength

The tensor strength was tested to investigate effect ofnanoparticles on mechanical characterization of the membranes.This test was done according to ASTM1922-03 and samples wereprepared in the standard shape in the ambient temperature (about26 8C) and humidity (28%). The result was reported as the averageof three tests for each sample.

3. Results and discussion

3.1. Water flux and Pb(II) rejection

In the first step we tried to optimize concentration of celluloseacetate (CA) as additive polymer in base polymer polyvinylchloride (PVC). What we expected from this addition was highwater flux, because the special feature of CA is its hydrophilicity[13] so various concentrations of CA (2.5, 5, 7.5, 10, 15, 25, 50, and75 wt%) were used to choose best concentration. The membraneshaving CA concentration more than 10% were broken at 5 barpressure; therefore, these were eliminated from all tests andfurther studies were done only on CA 2.5, 5, 7.5, and 10%concentrations.

Fig. 2 shows the water flux of membranes against volumepassed through membrane at different CA concentrations. WhenCA concentrations increase the pure water fluxes of membranesalso increase. We expected this behavior because of the specialfeature of CA [13]. Increased CA concentration in dope solutionincreases hydrophilic characteristic of membranes, so high amountof water is adsorbed by membranes [14] and more water can passthrough membranes.

As we know, usually with increasing flux, rejection decreasesbecause ions easily pass through membrane. Fig. 3 investigateshow rejection changes by increasing CA concentration. This figurereveals that by increasing CA, at first rejection does not show thecorresponding change but at 10% of CA the rejection improves. Thisis due to negative charge of CA [26], which increases the negativecharge of membranes and causes improvement of lead ionsadsorption of positive ion (Pb2+). As seen in Fig. 3, there was a quickdecline in rejection in the first 12 cc which then decreases slowly. Itseems this occurs when the adsorption sites in the membrane are

Fig. 4. Rejection verses flux to comparison and selection best concentration of

polymer matrix.

Fig. 5. The effect of Fe3O4 nanoparticles concentration (0.01, 0.1, and 1) on flux.

A. Gholami et al. / Journal of Industrial and Engineering Chemistry 20 (2014) 1517–15221520

filled up by positive ions; this behavior is already reported by otherresearchers too, when the adsorption sites are filled up by ions,other ions pass through membrane easily [27]. Fig. 4 presents themembrane containing 10 wt% CA in comparison with others, so infirst step membrane with 10% of CA is chosen as best membrane.

In second stage, Fe3O4 nanoparticles in three differentconcentrations (0.01, 0.1 and 1 wt%) added to best membrane infirst stage (PVC–CA 10 wt%).

Fig. 5 shows the effect of different concentrations of Fe3O4

nanoparticles on flux of membranes. Some researchers [28,29]reveal that nanoparticles cause some changes in membrane’sstructures and improve their porosities, what we saw in SEMpictures, Fig. 6(a–d), nanoparticles increase channels in mem-branes too, but in membrane with 1 wt% nanoparticles, flux declinein comparison by membrane with 0.1 wt% nanoparticles.

As you see in SEM figures nanoparticles fill up majority of holesand channels so the flux reduced. Similar behavior has been

Fig. 6. SEM images from cross section of (a) PVC/CA (9:1 w/w), (b) PVC/CA (9:1 w/w)–F

reported by Daraei et al., too [28]. Certainly some particles arewrapped by polymer and this reduces real efficiency of membrane.

Fig. 7 display the Salt Rejection against volume passed throughmembrane in different Fe3O4 nanoparticles concentrations. Whennanoparticles concentration increases the salt rejection alsoincreases. This behavior is because of nature of nanoparticles thatact as an adsorbent, so we expected this behavior but lastmembrane (PVC–CA 10 wt%) treats in different way from others. Asshown in Fig. 6(d) the increase of nanoparticles concentration ledto nanoparticles accumulation and agglomeration and createshunks. These hunks reduce effective area of particles, so rejectiondecreases. Similar behavior has been confirmed by Daraei andZheng researches [28,29].

As shown in Fig. 7 nanoparticles (0.01% and 0.1 wt%) caused100% increasing lead rejection at steady state condition and morethan 200% in unsteady state type. The reusability of themembranes was examined using EDTA as eluting agent in many

e 0.01 wt%, (c) PVC/CA (9:1 w/w)–Fe 0.1 wt% and (d) PVC/CA (9:1 w/w)–Fe 1 wt%.

Fig. 7. The effect of Fe3O4 nanoparticles concentration (0.01, 0.1, and 1) on lead

removal.

Fig. 8. The effect of Fe3O4 nanoparticles concentration (0.01, 0.1, and 1) on water

content.

Fig. 9. The effect of Fe3O4 nanoparticles concentration (0.01, 0.1, and 1) on tensor

strength (mechanical properties).

A. Gholami et al. / Journal of Industrial and Engineering Chemistry 20 (2014) 1517–1522 1521

researches. The result showed good performance for regeneratedmembrane [16].

3.2. Water content

Obtained results (Fig. 8) showed that membrane water contentincreased by increase of nanoparticles concentration up to 0.1 wt%in prepared membranes. The membrane water was decreasedagain by more increase in nanoparticles content to 1 wt% becauseof pores and channels filling by nanoparticle at high additiveconcentration which decrease the voids and cavities in membranematrix [28].

The swelling in prepared membranes was less than 5% inthickness. Moreover this was negligible in length and width. Theloss of molecules solvent in preparation stage causes voidsformation which are sufficient to accommodate the watermolecules. Hence, the solvation does not change membranedimensions manifestly.

3.3. Membrane tensor strength

As you see in Fig. 9 the mechanical strength of pristinemembrane (sample 1) and modified membrane containing0.01 wt% nanoparticles was almost same. As mentioned beforeincrease of nanoparticles concentration causes to more channelsformation in membranes matrix which decreases membranemechanical strength. Also channels filling by nanoparticles canincrease the mechanical strength abruptly.

4. Conclusion

Novelty of this study is using ferrosoferric oxide (Fe3O4)nanoparticle to remove Lead from water. A series of polyvinylchloride (PVC)–cellulose acetate (CA)–Ferrosoferric oxide (Fe3O4)Nano filtration membranes were prepared by phase inversionmethod. In first step different concentrations (2.5, 5, 7.5, 10, 15, 25,50, and 75 wt%) of CA were used as additive to improvehydrophilicity of membranes and finally 10% of CA was pickedas best concentration. Membrane containing 40% wt CA showedsuitable Lead rejection. In Second step different concentrations(0.01, 0.1 and 1% wt) of Fe3O4 were used to improve membranerejection. Membrane with 0.1% of Fe3O4 showed better flux andrejection compared to others.

Acknowledgment

The authors gratefully acknowledge Arak University for thefinancial support during this research.

References

[1] M. Hua, S. Zhang, B. Pan, W. Zhang, L. Lv, Q. Zhang, Journal of Hazardous Materials211–212 (2012) 317.

[2] J. Garcıa-Leston, J. Mendez, E. Pasaro, B. Laffon, Environment International 36(2010) 623.

[3] M. Amini, M. Arami, M.M. Niyaz, A. Akbari, Desalination 267 (2011) 107.[4] N. Ghaemi, S.S. Madaeni, A. Alizadeh, P. Daraei, F. Rahimpour, A.A. Zinatizadeh,

Separation and Purification Technology 85 (2012) 147.[5] V. Vatanpour, S.S. Madaeni, R. Moradian, S. Zinadini, B. Astinchap, Separation and

Purification Technology 90 (2012) 69.[6] S. Yua, M. Maa, J. Liua, J. Tao, M. Liua, C. Gao, Journal of Membrane Science 379

(2011) 164.[7] E.M. Vroenhoek, J.J. Waypa, Desalination 130 (2000) 265.[8] B.B. Tang, C. Zou, P. Wu, Journal of Membrane Science 365 (2010) 276.[9] H.S. Lee, S.J. Im, J.H. Kim, H.J. Kim, J.P. Kim, B.R. Min, Desalination 219 (2008) 48.

[10] E.S. Wiks, Industrial Polymers Handbook: Products, Processes, Application,Wiley-VCH Press, Germany, 2001.

[11] J.E. Mark, Polymer Data Handbook, Oxford University Press Inc., New York,1999 .

[12] C.A. Harper, Handbook of Plastic and Elastomers, McGraw-Hill, New York, 1975.[13] R. Abedini, S.M. Mousavi, R. Aminzadeh, Desalination 277 (2011) 40.[14] S.M. Hosseini, A. Gholami, S.S. Madaeni, A.R. Moghadassi, A.R. Hamidi, Desalina-

tion 306 (2012) 51.[15] L. Yan, Y.S. Li, C.B. Xiang, Polymer 46 (2005) 7701.[16] F. Liu, M.R. Moghareh Abed, K. Li, Journal of Membrane Science 366 (2011) 97.[17] M.A. Al-Anber, Desalination 250 (2010) 885.[18] I. Soroko, M.P. Lopes, A. Livingston, Journal of Membrane Science 381 (2011) 152.[19] I. Sorokoa, M. Makowskia, F. Spill, A. Livingston, Journal of Membrane Science 381

(2011) 163.[20] X.-Z. Wei, L.-P. Zhua, H.-Y. Denga, Y.-Y. Xua, B.-K. Zhu, Z.-M. Huang, Journal of

Membrane Science 323 (2008) 278.[21] A.Z.M. Badruddoza, A.S.H. Tay, P.Y. Tan, K. Hidajat, M.S. Uddin, Journal of Haz-

ardous Materials 185 (2011) 1177.[22] Z.-Q. Huang, K. Chen, S.-N. Li, X.-T. Yin, Z. Zhang, H.-T. Xu, Journal of Membrane

Science 315 (2008) 164.

A. Gholami et al. / Journal of Industrial and Engineering Chemistry 20 (2014) 1517–15221522

[23] X. Zhang, Y. Chen, A.H. Konsowa, X. Zhu, J.C. Crittenden, Separation and Purifica-tion Technology 70 (2009) 71.

[24] S.M. Hosseini, S.S. Madaeni, A.R. Heidari, A.R. Moghadassi, Desalination 279(2011) 306.

[25] J. Huang, K. Zhang, Desalination 282 (2011) 19–26.[26] N. Ghaemi, S.S. Madaeni, A. Alizadeh, P.D.V. Vatanpour, M. Falsafi, Desalination

290 (2012) 99.[27] S.S. Madaeni, F. Heidary, Applied Surface Science 257 (2011) 4870.

[28] P. Daraei, S.S. Madaeni, N. Ghaemi, E. Salehi, M.A. Khadivi, R. Moradian, B.Astinchap, Journal of Membrane Science 415–416 (2012) 250.

[29] Z.-H. Zhang, Q.-F. An, T. Liu, Y. Zhou, J.-W. Qian, C.-J. Gao, Desalination 297 (2012)59.

[30] S.M. Hosseini, S.S. Madaeni, A.R. Khodabakhshi, Preparation and characterizationof PC/SBR heterogeneous cation exchange membranes filled with carbon nano-tube, Journal of Membrane Science 362 (2010) 550–559.

[31] Y.H. Lia, J. Dinga, Z. Luan, Z. Di, Y. Zhu, C. Xu, D. Wu, B. Wei, Carbon 41 (2003) 2787.